No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Abstract
For the improvement of the operation speed and the fault-tolerant ability of the distribution network's fault location, a new algorithm is proposed in this paper: Based on the chain table model, branch is taken for the unit to be used for the first location and searching the fault branch at the first. Then the data inconsistency between the head node and the end node could be used for the second location according to the link relation of the nodes in the fault branch; Local reconstruction is used to treat the nodes without data; The data correctness of each node is judged by the first location or the data relationship of the nodes in each furcation, on which some fault-tolerant mechanisms that could deal with one or more points fault and their correct rates are discussed. Actual operation results and the comparision of the complexity between present algorithm and the traditional matrix algorithm verify the high effectiveness and good fault-tolerant ability of this algorithm.
... With the development of distribution automation systems and monitoring devices, DTU and FTU are gradually popularized, which can monitor the electrical status and alarm state during the operation of the distribution network and will timely upload the electrical information to the data center (SCADA) after fault detection. On this basis, some fault location methods are proposed for the active distribution network, such as expert system (ES) [3], artificial neural network (ANN) [4,5], Petri nets [6], the rough set method [7], linked-list method [8,9], matrix algorithm [10][11][12][13][14][15], and optimization algorithm [16][17][18][19][20][21][22][23][24][25][26][27][28][29][30]. Among them, the matrix algorithm and the optimization algorithm, are widely used in the practical distribution network. ...
... " + " indicates that the alarm information is distorted by 1, "−" indicates that the alarm information is distorted by −1, and " 0 " indicates that an alarm message is missing. 8 International Transactions on Electrical Energy Systems complementation of both algorithms' advantages improves the performance of the fault location. When there is no alarm information distortion, the improved matrix algorithm can quickly and effectively determine the real fault section without using the optimization link. ...
The traditional fault location methods are not effective for the multisource active distribution network and are difficult to meet both requirements of timeliness and fault tolerance simultaneously. In this paper, a fault location method for active distribution networks which combines both improved matrix algorithm and optimization algorithm is proposed. The improved matrix algorithm is constructed to determine the fault section hypothesis by using the causal relationship between the alarm information and the fault section. Subsequently, network splitting is used for causal verification to check whether the alarm information is distorted. If the alarm information is normal, the fault sections can be quickly located by the above matrix algorithm. If the alarm information is distorted, the optimization model is constructed by using the fault section hypothesis selected by the matrix algorithm. The discrete particle swarm optimization (DPSO) algorithm is used to solve this optimization model, which can accurately locate the real fault sections. Hence, the advantages of both matrix algorithm and optimization algorithm are complementary. Simulation results show that the proposed method has the advantages of both high timeliness and fault tolerance in a complex active distribution network.
... For a network with n nodes, the network infrastructure matrix is a square matrix of n rows and n columns representing the connection relationship between nodes Shi et al. (2014). Based on the characteristics of distribution network structural variants, this method can effectively represent the topology of distribution networks, but it is a matrix-based representation method, and the matrices of distribution networks are very sparse and occupy large storage space Wen et al. (2009). ...
The mathematical analysis and improvement of matrix algorithm based on FTU has been proposed to solve fault location problem in complex multi-source and multi fault distribution networks of power system. Based on the structural characteristics of the distribution system, a description matrix D is established by assuming the power sources and a positive direction and it derived from the correlation between the positive direction and different power lines. As a function of the fault current direction transmitted by the terminal power supply unit which is our feeder terminal unit (FTU), the fault information vector F is set up. By combining the description matrix D and searching for non-zero elements in the fault information vector, the fault location vector is discovered according to the fault location criteria. During the observation of the elements in the fault location vector, by analyzing we immediately found the location of the defect area. This algorithm can solve problems that cannot be completely solved by other algorithms, such as power terminal failures, ring network failures, and multi-source failures. This method does not require matrix multiplication or normalization. The widespread use of distributed generation (DG) has made the distribution network more complex, and it leads to the failure of the traditional matrix algorithm. Therefore, the matrix algorithm is further improved to adapt the complex characteristics of DG. Considering the accuracy, we avoid random search operations by filtering fault candidate scenarios based on fault confidence and Each algorithm run 100 times in a loop, and the average time taken for a single run is used as a measure of the computational efficiency of the algorithm. The matrix algorithm utilizes the information uploaded by FTU to SCADA to create network description matrix, fault information matrix, which are then used to obtain the fault judgment matrix for fault location and isolation and it proves that the algorithm judgment is effective.
... The artificial intelligent fault diagnosis scheme is rising in recent years, and is able to position and diagnose faults based on non-sound information. Its relevant theories have expert system, artificial neural network, fuzzy theory, genetic algorithm and immune algorithm etc. [5][6][7][8][9]. Since most FTUs are located outdoors and heavily affected by environments, it is more likely to get the wrong information using the matrix algorithm. ...
Fault location is an effective way to ensure quickly power recovery after a failure in power distribution system. Bad operation environment of FTU, damaged elements or lost information usually cause variations in fault information for correlation analysis in fault location of power distribution system. A data mining (DM) correlation analysis model based on rough sets (RS) theory and immune algorithm (IA) is proposed in this paper. Firstly, using RS theory to extract domain knowledge, a set of the given variant fault pattern is converted into a decision table in RS. Secondly, an attribute reduction of the decision table is made by using the IA theory, and the intrinsic correlation rules between input vector (condition attribute) and output vector (decision attribute) are mined. Then, the data mining method is used to deal with the distortion of FTU real-time input information. According to the current limit information sequence of section switches, line fault states in each section are judged to realize the fault location in power distribution network. Finally, the feasibility and effectiveness of the fault location model of distribution network based on RS-IA data mining model is verified by simulation.
... Paper [6] proposes methods for positioning the feeder terminal area fault and multiple source multiple faults, but the discriminant theorem is too complicated for its need of a discriminate formula to get fault region after obtaining fault interval judgment matrix. Based on matrix algorithm and according to tree structures of the distribution network, paper [7] uses chain table method. It raises the positioning speed and can judge multiple faults even under the condition of distortion. ...
An improved FTU-based matrix algorithm is presented for faults locating in complex distribution networks with multiple source multiple and faults. Firstly, the paper supposes the only power source and the positive direction according to the structural characteristics of the distribution system, and build the description matrix D which is based on the positive direction and the relevancy between different feeders. Based on the fault currents direction sent from the feeder terminal unit (FTU), we can set up the fault information vector F. Secondly, combined with the description matrix D, by searching the nonzero elements in the fault information vector F , we can establish the fault location vector L according to the fault locating criterion. Finally, by observing the elements in fault location vector L, we can find out the fault area location quickly. This algorithm can solve the problems of feeder terminal fault, circle net fault and multiple source multiple fault, all of which cannot be solved by other algorithm completely. And this method needs not matrix multiply or standardization. It's easy to build the description matrix D, which is very sparse and has good real-time and practicality. The criterion of this algorithm is proved to be effective by simulation in MATLAB environment.
In order to deal with the possible loss and distortion of the information from feeder terminal units (FTU) causing the incorrect fault location in distribution networks, a new method based on information restoration preprocessing with information replenishment and correction as well as membrane computing is presented. The incomplete information from FTU is replenished and the distortional information is detected and corrected by using an information restoration preprocessing model so that the real information describing the over current can be obtained. Furthermore, by employing the completed and corrected information, a new membrane computing with high precision by using fishermen optimization algorithm (FSOA) in the local optimization and rapid convergence by using ABC in global optimization is introduced to fault section location to improve the diagnosis accuracy. Compared with genetic algorithm (GA) and Electromagnetism-like mechanism (ELM), the proposed method shows distinct advantages in accuracy, speed and tolerance.
Distributed generation (DG) has been largely used in distribution network in recent years, which brings more security, reliability, and economy, but some bad effects to the distribution network. In this study, recovering more loads, less switching actions and distribution network power loss are considered to establish the mathematical model of fault recovery in distribution network with DG. The particle swarm optimisation algorithm is used to solve the problem, and the fault recovery strategy is obtained. Besides, the impact of DG location on the fault recovery of distribution network is studied, and the simulation results show that when the DG is located dispersedly in the downstream of the distribution network, the distribution network recovery is in good condition. The results of this study provide a reference for the DG configuration of the distribution network.
To solve the premature issue of particle swarm optimization (PSO), this paper introduces the idea of simulated annealing into the dynamic inertia weight adjustment to make the inertia weight decrease with the increase in the number of iterations and avoid the algorithm falling into local optimization. And it uses population fitness variance to judge the particles' similarity in order to mutate the particle adaptively, i.e. control the mutation number "big then small" and select particles randomly and rebuild displacement. In order to improve the search speed of PSO algorithm, the PSO algorithm and branch exchange algorithm are combined, in which the particle displacement is just the contact switch connecting non-fault power-lossing area and normal area, as for the scheme of forming looped network, the branch exchange algorithm is used to determine the section switch to open. The combination of PSO and branch exchange could ensure the radial operation requirements of distribution grid, and make the network formed by a lower net loss. The example shows that the distribution network fault recovery based on branch exchange-PSO can quickly restore power supply of power failure area.
In order to solve the fault section location problem of single-phase-to-ground fault in digitalized distribution grid with arc suppression coil in parallel/series with resistance, the fault zero-sequence currents measured by protection devices installed at the head ends of all sections in such distribution grid are formulated, and the general distribution regularities of the active component of zero-sequence current are provided. On this basis, the zero-sequence active component-based single-phase-to-ground fault section location method is proposed with the combination of general matrix algorithm for fault location in distribution grid. Moreover, the corresponding logic node and logic device is newly developed according to IEC61850, which provides reference to the device model construction of fault section location system of single-phase-to-ground fault in compensated network. The correctness of the proposed algorithm is verified by the simulation results of several typical faults in a distribution grid model built in PSCAD/EMTDC.
As distributed generations in distribution networks have brought much influence to the fault current, the traditional fault-section location method of distribution networks may not work. By building a new switch function, this paper puts forward a new fault-section location method. The proposed method is based on a binary hybrid algorithm and can adapt itself to the change caused by the switching of DG. The positive direction needs to be determined only once, when the proposed method is used to locate multiple fault sections in multi-source distribution networks. The binary hybrid algorithm which is a combination of particle swarm optimization and differential evolution is based on a two-population evolution scheme and an information exchange mechanism. The simulation results indicate that the proposed method has accurate fault-section location abilities when it is used to locate single fault or multiple faults. In addition, the fault-tolerance and efficiency of the proposed method is verified by the results.
In order to improve the practicality of feeder automation, this paper adopts the hierarchical and divisional configuration scheme of feeder automation according to features of the distribution network. The feeder automation is divided into three layers from bottom to up: feeder layer, feeder island layer and system layer. On the feeder layer and the feeder island layer, different feeder automation schemes are suggested based on the regional characteristics. The FA mode without communication which is simple and reliable is adopted in the single power line. This paper combines the reclosing the breaker twice with opening the loader breaker to isolate the fault of the feeder's branch line. A distributive and intelligent feeder automation method based on the sum of fault signals is proposed, which is simple and intuitionistic. This method is able to transform the judgment logic automatically between contact switch and section switch and adapt to the adjustment of the operating mode flexibly. On the system layer, this paper uses network topology and optimization operating to improve the reliability of power supply when the failed main network causes widespread power outage through the power supply of distribution lines. Furthermore, this method can optimize the network operating and reduce the power outage.
In order to realize robust fault location and isolation of distribution systems under the condition of incomplete information, this paper proposes using the relevance among each phase of fault information, three-phase fault currents and both reported by each feeder switch after fault to amend the omitted and false fault information reported by terminal equipment. And then, a method is proposed to achieve rapid fault location of distribution system's fault section by analyzing the topology, dividing the system into independent minimum distribution area and combining the corrected fault information reported by the above minimum distribution area. The simulation results show that the algorithm can effectively correct the missing and false fault information in order to achieve rapid fault-tolerant location of distribution system's fault section.
A new method of fault section location and isolation in distribution networks based on immune algorithm (IA) is presented. Immune algorithm gets individual evaluation from its affinity. Selection of the crossover and mutation individuals is not only based on its affinity but also concentration, which increases colony diversity, avoids partial optimization, and speeds up the rate of evolution. Besides, this method can establish memory units and retain part of the optimal solutions in order to avoid degeneration in the process of crossover and mutation and guarantee the result is the individual of the highest level of affinity when the algorithm is terminated. A method based on the idea of area division and fault alarm of the incoming break is proposed to improve the speed of location for open-loop operational distribution networks. Finally the simulation results show that the immune algorithm can realize the goal of fault location and the comparison results with genetic algorithm testify its validity and reliability.
To study the distribution network multi-fault rush repair problem with multiple repair squads, rational allocation of the repair tasks is the premise to complete the repair work smoothly and efficiently before the repair sequence optimization. Considering both the multiple repair squads with different initial positions and load itself failure, this paper defines such attribute vectors as the repair preference and ability vector of the repair squad and the fault demand vector, based on which the utility function model that can solve the collaboration problem is built and the utility function matrix is formed. Then via the roulette wheel selection method, the task allocation scheme set based on optimum utility is obtained. By quantizing the repair information in the form of the attribute vector and changing the attribute vector dimension, the uncertainty problem of the repair resources and fault demand can be resolved and realizing the dynamic allocation. Ten allocation schemes are given in the simulation result and the optimal one is obtained via the improved bacterial colony chemotaxis (BCC) algorithm, together with its corresponding optimal repair sequence, which proved the effectiveness and correctness of the proposed method.
This paper proposes a novel fault location method for electrical distribution systems based on characteristic frequencies, which uses solely the measured voltage transients without multi-recording nodes. Fault-generated traveling waves propagate along the distribution network and are reflected at the discontinuity points, so the recorded transients contain some characteristic frequencies associated with specific paths. Firstly, the recorded transients are analyzed by fast Fourier transform (FFT). Secondly, the sub-frequency bands are identified and the energy values of all sub-frequency bands are figured out. Then the fault position can be located based on the energy values. PSCAD/EMTDC and MATLAB are used for the simulation validation work and the sensitivity and the influencing factors of this method are analyzed. The simulation results have proved the effectiveness of the method.
In order to deal with the possible loss and distortion of the information from feeder terminal units (FTU) causing the incorrect fault location in distribution networks, a new method based on information restoration preprocessing with information replenishment and correction as well as membrane computing is presented. The incomplete information from FTU is replenished and the distortional information is detected and corrected by using an information restoration preprocessing model so that the real information describing the over current can be obtained. Furthermore, by employing the completed and corrected information, a new membrane computing with high precision by using fishermen optimization algorithm (FSOA) in the local optimization and rapid convergence by using ABC in global optimization is introduced to fault section location to improve the diagnosis accuracy. Compared with genetic algorithm (GA) and Electromagnetism-like mechanism (ELM), the proposed method shows distinct advantages in accuracy, speed and tolerance.
A survey of fault location technology for distribution networks is introduced based on existing research results, including faulty line selection, faulty section location and fault distance detection. The faulty line selection is used to distinguish the faulty one among multiple feeders and it is mainly based on the additional signal injection, the fault fundamental component, the fault transient component or the synthetic signals. The faulty section location is used to quickly isolate the faulty area for power recovery and it is mainly based on the real-time fault information uploaded from field devices, the trouble call or the weak fault characteristic detection. The fault distance detection is used to directly determine the fault location and it is mainly based on the signal injection, the fault fundamental component or the fault transient component. Combined with the key technologies of automatic fault location and its latest research achievements, its development trend is prospected.
The optimization strategy for wide range of load transfer and its implementation in distribution networks during power supply side failure is the extension of feeder automation. It will be of great significance to improve the reliability of distribution network. Firstly, the paper defines the types of busbar voltage loss. Thus the problem of large scope blackout is decomposed into different types of busbar voltage loss problems. After that the optimization strategy for wide range of load transfer of different types of busbar voltage loss is proposed, where the wide range of load transfer problem is transformed into the problems of fault recovery of equivalent feeders and load transfer within the substation. Finally, a scheme of a wide range of load transfer is investigated and the flowchart of each module is given. This research provides a method and technical support for the application of wide range of load transfer under busbar voltage loss in distribution networks.
With the increasing interconnection of distributed generators in distribution networks, some of the existing fault location algorithms may fail. A fault location method based on the minimum 2-norm of voltage deviation vectors is proposed. Firstly, the fault current is injected into each bus of the distribution network model to calculate the voltage variation of the power source. Secondly, the difference between measured voltage variation and calculated voltage variation is analyzed, and the fault current injected bus corresponding to the minimum difference is searched and determined as the first bus of the fault section. Finally, the other bus of the fault section is determined according to the variation trend. An IEEE 34-bus system is chosen to verify the algorithm, the results showing that the algorithm has clear physical meaning, simple calculation process and good fault location accuracy. The more distributed generation is connected, the better fault location accuracy will be.
Based on the chain table model, a new algorithm is proposed to solve the issue of on-line location of single phase grounding fault of complex distribution network. Firstly, the data of fault information is obtained by FTU. Then taking branch for the unit and using the data between the head node and the end node, the fault branch is found in the first location. Finally, according to the information at the two ending nodes of each region, accurate location is realized in the second location along the indicated direction. Some particular cases, such as the failure between the branches, the scarcity of nodes information and the distortion of the data measured by FTU, are analyzed and corresponding methods are discussed. The algorithm shows characteristics of simple principle, accurate location, high speed calculation and better fault-tolerance, making on-line locating of the grounding fault a reality. Through a case analysis, the correctness and feasibility are verified.
The shortcomings of current fault location algorithms for distribution network are analyzed; network relationship matrix and modified matrix algorithm for fault location are proposed combined with the structure characteristics of distribution network. Comparing to the matrix algorithm and its improved algorithm which are proposed before, this algorithm has the feature of simple principle, less computation without complex computing, good real-time performance. It can be applied to both simple distribution network with single power source and complex distribution network with multiple power sources, not only the single point fault but also the multiple point faults can be located quickly and accurately. Even the signals which FTU on pillars send to master station distort, fault point still can be located quickly and accurately, it has very good fault tolerance. Analyses and simulations demonstrate the feasibility and accuracy of the modified matrix algorithm.
The fault sections location in distribution networks based on the information of the short circuit current may get false results due to the information aberrance. To solve the problem, firstly this paper classifies the information aberrance, then proposes the information contradiction theory based on the matrix algorithm to detect recognizable aberrance, analyzes the possibility of aberrant node through the Bayes theorem, and proposes the scheme of aberrance correction and fault location. This strategy can find the position of aberrant node quickly, only extract part of the fault information for correcting aberrance, and for the complex aberrance, expand fault segment appropriately to ensure the reliability of fault location. The strategy has accurate location and fast speed and high reliability. It can solve the fault location of multiple faults and multiple information aberrance in complex multiple power system.
Fault indicator is widely used in distribution power network. Its number and installation location both affect the fault interruption time. This paper takes use of the immune algorithm to study the optimal installation number and installation location of fault indicator in distribution network so as to achieve the maximal benefit.
Currentlysingle-phase ground protection in coalmine high-voltage distribution network still relies on time stage difference coordination to realize the longitudinal selectivitywhich poses a potential threat to power supply security.Thereforedistribution characteristics of both the active component of zero-sequence current fundamental wave and the reactive component of zero-sequence current harmonic component in the coalmine distribution network were analyzed and elaboratedmeanwhile Rogowski coil was introduced into the detection of zero-sequence currentthe analysis of the amplitude and phase angle characters of secondary voltage signal generated in measurement of the zero-sequence current using Rogowski coil shows that the percentage of zero-sequence current harmonic component in the current being measured can be effectively amplified when measured by Rogowski coil.On this basisthe criterion of synthetic zero-sequence current based on the secondary signal of Rogowski coil is proposedand the algorithm of single-phase ground fault section location in coalmine high-voltage distribution network is presented.This algorithm not only achieves location of single-phase ground fault section without stage difference coordinationbut also effectively improves the sensitivity of the single-phase ground protection.The correctness of the proposed algorithm is verified by typical single-phase grounding fault simulation in 10 kV coalmines high-voltage distribution network system in PSCAD/EMTDC.
The premise of effective use of clean energy is to correctly determine malfunction zone of distribution network with distributed generation. Based on the matrix algorithm in the location of feeder terminal unit (FTU) to the malfunction of distribution network, a new adaptive matrix algorithm for the fault location in distribution network is proposed. According to flowed over-current through the FTU and its direction, firstly the algorithm judges whether distributed power put into operation, and whether the default happens. Then it preliminarily locates the malfunction zone in the distribution network. Secondly, on the basis of malfunction zone structure and over-current information of FTU, the fault matrix is formed adaptive, we could confirm of the fault zone. Lastly, comparing the detected fault current amplitude at both ends of fault circuit to form fault criterion, it could effectively exclude effects of incomplete and fault distortion information. The PSCAD/EMTDC simulation model and numerical example verifies accuracy and efficiency of the algorithm, and good mechanism of fault tolerance. Key words-distributed generation; distribution network; FTU; fault location; adaptive matrix algorithm
Although good application effect of locating faulty section in complex distribution network can be attained by applying ant colony algorithm, however there are also disadvantages such as too long search time and computation time. For this reason, traditional ant colony algorithm should be improved. Firstly, a switching function that can dynamically adapt to topological structure of distribution network is constructed; then a method to set the initial values of ant colony's pheromone and the disturbance rule of the solution are led in; finally the method to apply the improved ant colony in fault section location of distribution network with multi power supplies is determined. The advantages of the improved ant colony algorithm in calculation speed and error tolerance are verified by results of calculation example.
As the complexity and changeable characteristic of the topology structure, the particularity of the dynamic discrete event description during fault in the DS (distribution systems), the application of PN (Petri nets) in the FSL (fault section locating) is restricted. In this paper, a new approach of the FSL for DS is proposed, which is developed based on Petri Nets in combination with redundant correcting technique. First the fault topology structure is built adaptively based on the redundant correcting technique through the information analysis of SCADA system from FTU (feeder terminal unit). The redundant correcting technique is used for the correcting information distorted in the real-time fault information. Secondly according to the electrical measure in fault topology structure, the terminal FTU in fault topology is determinate and locked which is used as building Petri nets for the solving of fault section. So the problem is changed to solve the FSL on the locking FTU node. The approach proposed has better generality, simplicity, fast and exact solving. It is effective to enhance the fault tolerant performance of FSL.
Based on rough set theory, the paper proposed an new distribution network fault diagnosis approach to deal with the imperfect alarm signals, caused by malfunction or failing operation of protection relays and circuit breakers, error in the communication equipment. Due to rough set theory can effectively handle the imprecise problems without any ancestor information except the data set itself, a decision table including all kinds of fault cases is established by considering the signals of protection relays and circuit breakers. Then, diagnostic rules are extracted by reducing the decision table. Using the predicts of decision table, the diagnosis rules can be obtained directly from fault samples which have been established. The method can tell indispensable fault signals from the dispensable ones and discover the inherent redundancy of alarm signal set. Finally, a practical fault diagnosis program is proposed based on VC++ 6.0, with the main interface made by VB6.0. A lot of simulation results show that the method is quite efficient and has excellent fault-tolerance capability.
In order to promote the calculation efficiency and expand its adaptability to the complex power distribution system, an improved and optimized matrix algorithm, which is a fault section locating method depended on local information of power distribution system, is proposed for distribution network according to the model in combination with the fault information. Firstly, the algorithm needs to form the network description matrix, then the local fault information of power distribution to gain the judgment matrix and the situation of faults can be obtained by the analysis of the judgment matrix, this method suits to radial or mesh architecture in radial operation distribution systems. The algorithm is designed based on fault overcorrect signal flowing through monitoring points, and parameters of the distribution are all-stored according to the power line and the net describing matrix is dynamic to be formed. By the algorithm not only the single fault occurred in distribution network can be located, but also the end faults and the multiple faults can be quickly and accurately diagnosed. With the help of efficient database, the simulation and calculation of various faults occurred in a distribution network with three independent sources and the faults occurred in a distribution network consisting of single source and multi-outgoing lines are carried out, the result indicates the algorithm is effective. There is no need for matrix multiplication and reversal calculations in the improved algorithm, therefore the method could find its application in on-line calculation because of its high-speed character. Furthermore the method may also be extended to systems with multi-power sources and provides adaptability to different open-loop power distribution networks.
The study based on rough set theory (RST) and data mining (DM) is used for fault section diagnosis (FSD) of distribution networks. The four kinds of diagnosis models on the basis of the different combination RS with neural network (NN) are proposed. The different complementarity and associated relation, the different application mechanism and principle, the model limitation are presented by implementing RS and NN in four models and the all-around assessment of the application mechanism and performance of RS is given. Furthermore, comparisons of simulation results among the five kinds of models prove the significance of DMs discovery in hidden knowledge. The conclusion and assessment of each model have the same significance for fault diagnosis system in other fields.
C-E (cause and effect) nets model has a superiority in reaching diagnostic result quickly. However it is affected by the uncertainty in the available information because of the reasoning properties of shallow knowledge. Because the local area communication in substation can obtain omnibearing information, such as the SOE, switching information sequence of circuit breakers, fault recorder information, this paper proposes an on-line diagnostic approach of hybrid C-E nets model in substations. The essence of the approach is that the redundant information is extracted on the basis of using sufficiently the correlation between information and led in redundant correcting analysis and deep knowledge reasoning. Five types of nodes of C-E information relation are presented to build hybrid C-E nets and its rule matrix. The key merit of proposed method is its ability to give accurate Boolean value for the state vector in C-E nets model. The simulation results show the effectiveness to enhance fault-tolerance performance of fault diagnosis and the high applicability of the hybrid C-E nets model.
Based on Bayes probability likelihood ratio, a new fault location algorithm for distribution networks is proposed. A fault diagnosis model is established firstly, and then by using the fault information sequence gathered by the SCADA system from load switches, the probability likelihood ratios are calculated using the algorithm proposed. Finally, by searching the maximal fault probability, the fault section in distribution networks can be identified and isolated, in which the error and aberrance of SCADA information are screened to a certain extent. As a practical method, the SCADA information based fault location method can quickly determine the fault location. Examples are given to verify its correctness and effectiveness.
The reaction of nitrite with ferric and ferrous porphyrins was examined using visible, infrared and NMR spectroscopy. Solutions of either ferric or ferrous porphyrin were stable in the presence of nitrite, with only complexation reactions being observed. Under voltammetric conditions, though, a rapid reaction between nitrite and iron porphyrins was observed to form the nitrosyl complex, Fe(P)(NO), where P=porphyrin. The products of the reduction of ferric porphyrins in the presence of nitrite were confirmed by visible spectroelectrochemistry to be Fe(P)(NO) and [Fe(P)]2O. Visible, NMR and infrared spectroscopy were used to rule out the formation of Fe(P)(NO) by the iron-catalyzed disproportionation of nitrite. A reaction between iron porphyrins and nitrite only occurred by the presence of both oxidation states (ferric/ferrous). The kinetics of the reaction were monitored by visible spectroscopy, and the reaction was found to be first-order with respect to Fe(OEP)(Cl) and Fe(OEP). The products were the same as those observed in the spectroelectrochemical experiment. The rate was not strongly dependent upon the concentration of nitrite, indicating that the coordinated, not the free nitrite, was the reaction species. The kinetics observed were consistent with a mixed oxidation state nitrite-bridged intermediate, which carried out the oxygen transfer reaction from nitrite to the iron porphyrin. The effect of nitrite coordination on the reaction rate was examined.